Need help selecting photodetector

[runarh]

The photodetector is to be used to detect green light from an optical fiber (emitted by fluorecein, about 500nm). I have tried using this detector, but with no readings. This diode has a response of about 0.25 A/W at 500nm.

This is the amplifier I have used (OPA27GP):

Is there a better amplifier for this, or do I need a more sensitive detector?
A detector with SMA905 mount for the fiber is preferred.

 

[ronsimpson]

What amplifier did you use? We need more information. Did you make the amplifier?

 

[runarh]

What amplifier did you use? We need more information. Did you make the amplifier?

I used an OPA27. The light from the fiber is not strong enough to be detected by my current setup.

 

[jpanhalt]

How easy is the light to see in a normally lit room?
John

 

[ronsimpson]

I used an OPA27. The light from the fiber is not strong enough to be detected by my current setup.

Again did you make the amplifier? What power supply voltage? What is the voltage on the output? More information!

Have you tried a light meter?

 

[runarh]

How easy is the light to see in a normally lit room?
John

Looking at the end of the fiber, it's visible, but not very. If I point the other end of the fiber towards the lights in the ceiling, this light is much brighter through the fiber than the green light.

Againid you make the amplifier? What power supply voltage? What is the voltage on the output? More information!

Have you tried a light meter?

I made the amplifier with +/- 15V supply. Output is 0V when measuring the green light through the fiber. Without the fiber (only ceiling light) the output is 0.3V. I do not have a light meter, but might be able to borrow one.

 

[jpanhalt]

Looking at the end of the fiber, it's visible, but not very. If I point the other end of the fiber towards the lights in the ceiling, this light is much brighter through the fiber than the green light.

That is what I thought might be the case. Fluorescence intensity can be misleading, because it is usually viewed in a darkened room. I suspect your actual intensity is below the sensitivity of the device you are using. We used photomultipliers for our quantitative fluorescence work.

John

 

[spec]

The photodetector is to be used to detect green light from an optical fiber (emitted by fluorecein, about 500nm). I have tried using this detector, but with no readings. This diode has a response of about 0.25 A/W at 500nm.

This is the amplifier I have used (OPA27GP):

Is there a better amplifier for this, or do I need a more sensitive detector?
A detector with SMA905 mount for the fiber is preferred.

Hi again run,

Afraid the amp circuit above is not suitable for the Edmund 62-270 optodetector diode. The OPA27GB opamp should be suitable though, depending on your maximum data rate.

I am giving your requirement some thought.

In the meantime, if you haven't already got it, here is a link to the Edmund 2016 Electrooptic Catalogue (intresting reading, but by God, is it big. ): https://cdn.coverstand.com/30093/285675/1f83b097802e75213e8078a307a916c72d8c071b.1.pdf

And here is some of the electrical design data for the 62-270 extracted from the catalogue:

​

Just a few questions:
(1) What is your preferred supply voltage for the optoreceiver amp (the transmitter is using 12V)?
(2) What maximum data rate do you need
(3) What output would you like? Would 5V, 74xxx logic levels be OK?

​

spec

 

[runarh]

​

Just a few questions:
(1) What is your preferred supply voltage for the optoreceiver amp (the transmitter is using 12V)?
(2) What maximum data rate do you need
(3) What output would you like? Would 5V, 74xxx logic levels be OK?

​

spec

1. Preferably +/- 15V from the myDAQ (unless it needs more current than the 2mA it delivers), or the power supply in the LED driver (12V).
2. Max 1kHz
3. Yes, that would be good.

 

[Les Jones]

Also the detector is not suitable for green light at 500nm. It's sensitivity is quoted at 950 nm (Infra red.)

Les.

 

[spec]

1. Preferably +/- 15V from the myDAQ (unless it needs more current than the 2mA it delivers), or the power supply in the LED driver (12V).
2. Max 1kHz
3. Yes, that would be good.

Excellent! will assume 12V for the opto detector PSU; current not expected to exceed 100ma

spec

 

[spec]

Also the detector is not suitable for green light at 500nm. It's sensitivity is quoted at 950 nm (Infra red.)

Les.

Hi Les,

It is quite true that the Edmund 62-270 has a peak output at around 950nm (green curve on graph below) but it still has a current output at 500nm and I am hoping this will be sufficient to generate a signal from. Or have I missed something?

​

spec

 

[runarh]

I'm aware of the peak at 950, but I hope the .25 A/W will be enough to avoid replacing the detector.

 

[Les Jones]

Hi Spec,
I should have read the spec properly. As you say there is not a great loss of sensitivity for green light. It does also say that it is designed for medium light levels. One for low light levels may be better. I do not understand the significance of the four graphs unless there are four different versions of the detector. The red line marked UV is the only one that shows any output at UV wavelengths. I prefer to be proved wrong than trying to get some OP's to answer question so you can solve their problems.

Les.

 

[spec]

Hi Spec,
I should have read the spec properly. As you say there is not a great loss of sensitivity for green light. It does also say that it is designed for medium light levels. One for low light levels may be better. I do not understand the significance of the four graphs unless there are four different versions of the detector. The red line marked UV is the only one that shows any output at UV wavelengths. I prefer to be proved wrong than trying to get some OP's to answer question so you can solve their problems.

Les.

Hy Les,

Yes, the colored lines on the graph threw me at first- whoever chose colored lines must have been a bit sadistic. You are right; the various colored lines on the graph relate to individual detector diode types from the Edmund range.

spec

 

[spec]

I'm aware of the peak at 950, but I hope the .25 A/W will be enough to avoid replacing the detector.

Don't worry run,

I am fairly sure that your optodiode will be fine at 500nm. Unfortunately, I haven't had time to investigate further but I will be back on it soon.

spec

 

[runarh]

Ok, thanks!

 

[large_ghostman]

I have used these they are good in natural light https://www.silabs.com/products/sensors/infraredsensors/Pages/Si1120.aspx I got a few of them from badger boards

Datasheet here https://www.silabs.com/Support Documents/TechnicalDocs/Si1120.pdf

 

[Tony Stewart]

I've used this Photo sensor from Panasonic.. $3
Not only is it great for green, but it is optically corrected to match the eye spectral response for camera use.
It works down to a few Lux and up to bright sunlight with a log scale.
https://www.farnell.com/datasheets/1809963.pdf but response time is in milliseconds

What is your BW Specification?

You can use it to make a LUX meter.

Any PD ought to work for green depending on Lux level

But UV is a different requirement.

 

[spec]

Hy run,

Here is a circuit for an opto receiver:

Circuit Description
(1) Ideally, the Opto Receiving Diode (ORD) should operate with a 50V reverse voltage for the best on-current gain, but I am hoping there is enough on-current with only 10V reverse voltage. I have also done a circuit of a voltage multiplier to generate around 50V, so that can be used if necessary.
(2) The first opamp (N1) is a TransImpedance Amplifier (TIA) which converts the ORD on-current into a negative voltage (less positive). The current-to-voltage conversion factor of the TIA is, -(R3*Ion) Volts.
(4) The output from the TIA is then fed into a Comparator Schmidt Trigger (CST) (N2) to provide a nice square wave at its output which swings between 0V and 12V. I haven't bothered to convert to TTL, but that would be no big deal.
(5) With no ORD on-current, the TIA output sits around 2V. The voltage reference point for the CST is around 1V, so the CST output will be at 0V.
(6) When the TCA output drops below 1V, corresponding to an ORD on-current of over 1 uA, the CST will flip over so that its output will be 12V.
(7) I have assumed a opto receiving diode on-current of around 2 uA, but this may not be the case, so the gain of the trans-conductance amplifier will probably need to be adjusted by altering the value of R3. The criteria is that when the ORD on current is present the TIA will try to achieve around -1V. It wont be able to achieve this because the TIA output will hit 0V first.
(8) R5 and R6 have no play in the main circuit function and are only there to provide a degree of isolation and help keep the opamps happy in the frequency domain.
(9) C10 will probably not be needed in view of the low baud rate of IKHz but, for higher data rates, C10 would need to be optimized for circuit layout and ORD/housing parasitic capacitances.
(10) Unfortunately, the OPA192 is only available in surface mount. The schematic pin-out is for a SOIC 8 (SO 8) pak (OPA192IDR). The OPA192 costs around $2 US or in England, £2 UK. There are also dual (OPA2192) and quad (OPA4192) versions. https://www.mouser.co.uk/search/refine.aspx?Ntk=P_MarCom&Ntt=152100822.

Incidentally, I have just posted an article covering the OPAx192: https://www.electro-tech-online.com/articles/game-changer-opamp-opax192.768/

spec

​

Datasheets
https://www.ti.com/lit/ds/symlink/opa192.pdf